Skip to content

Latest commit

 

History

History
63 lines (51 loc) · 2.71 KB

README.md

File metadata and controls

63 lines (51 loc) · 2.71 KB

Finite Transform Library (FTL)

FTL Logo

Introduction

Fast Than Light Transforms which provide Radon and Number Theoretic Transforms within the geometry of the Discrete Fourier Transform (DFT)

The library has two implementations. The original C implementation and the newer Python Implementation.

C implementation features:

  • NTTW Sub Library for Number Theoretic Transforms, high resolution (microsecond) timing, basic array and imaging
  • FRTW Sub Library for Discrete Radon Transforms (DRTs), Mojette Transforms, Discrete Fourier Transforms (DFTs) via FFTW

Python implementation features:

  • Initial Python implementations can be found in the scripts/ folder.

Features

This library provides C and Python implementations of the following transforms:

  • Number Theoretic Transforms
  • High Resolution Timing
  • PGM Imaging
  • Discrete Radon Transforms
  • Rader Transforms
  • Mojette Transform
  • Ghosts
  • Discrete Fourier Slice Theorem
  • Discrete Radon Transforms

Fast Fourier Transforms are available via the FFTW library.

Build

See the build page for details. Usually involves the following:

mkdir build
cd build
ccmake .. OR cmake-gui
make OR ninja

Scientific Publications

Chandra, S. S.; Svalbe, I.; Guedon, J.; Kingston, A. & Normand, N.
"Recovering Missing Slices of the Discrete Fourier Transform Using Ghosts"
Image Processing, IEEE Transactions on, 2012, 21, 4431 -4441
DOI: 10.1109/TIP.2012.2206033
FTL Module: FRTW/ghosts

Chandra, S. S.; Normand, N.; Kingston, A.; Guedon, J.; Svalbe, I.
"Robust digital image reconstruction via the discrete Fourier slice theorem"
Signal Processing Letters, IEEE, 2014, 21(6), 682-686
DOI: 10.1109/LSP.2014.2313341
FTL Module: FRTW/mojette

Chandra, S. S.; Svalbe, I.
"Exact image representation via a number-theoretic Radon transform"
Computer Vision, IET, 2014, 8(4), 338-346.
DOI: 10.1049/iet-cvi.2013.0101
FTL Module: FRTW/radon

Chandra, S. S. "Fast Digital Convolutions using Bit-Shifts FTL Module: Coming soon

Licensing

This library is available under the terms of the Lesser GNU Public License v3.